Changes in the health care system are transforming existing models of cardiovascular disease management. Despite the disruption they will cause, these forces can potentially foster a new approach to cardiovascular disease that places prevention at the heart of patient-focused clinical practice. Harnessing these changes is not inevitable. Current mechanisms for reengineering care to meet these challenges are inadequate. This proposal seeks to train a young physician in the new sciences of change management. Through this award, Randall S. Stafford, MD, PhD will build on his experience in internal medicine, epidemiology and health care management to develop a research career committed to improving the quality of health care. The aim of the proposed research is to use a data-driven, quantitative approach to design, implement and evaluate quality improvement efforts. It uniquely pairs analysis of national data with local clinical improvement initiatives. Data from the 1973-95 National Ambulatory Medical Care Surveys and the 1991-94 Medicare Current Beneficiary Surveys will be analyzed. This investigation will focus on cardiovascular disease prevention and treatment. Multivariate statistical analysis will test research hypotheses addressing: 1) the adequacy of current practices, 2) the role of race, gender, health insurance and physician specialty in practice patterns, and 3) the effectiveness of clinical guidelines. These results will be applied directly to local quality improvement initiatives. Local projects will be better able to prioritize problems, confront obstacles to quality practice, and identify methods of influencing physicians. The research environment for this proposal will encompass the Partners Health Care System (Massachusetts General Hospital [MGH] and Brigham and Women's Hospital). Research analysis will be carried out within the MGH General Internal Medicine Unit. Clinical improvement projects will include: 1) an initiative to increase the use of anticoagulation in atrial fibrillation, 2) enhancement of hospital critical pathways to emphasize prevention, 3) development and use of quality indicators for outpatient cardiovascular disease practices, and 4) evaluation of community-oriented initiatives to reduce ischemic heart disease risk in Charlestown, MA. These natural experiments will allow a rigorous examination of the effectiveness of quality improvement efforts. The most promising initiatives will form model systems that can be disseminated to other health care institutions to improve the quality of cardiovascular care.